🌏 IndicVerse

IndicVerse is dedicated to advancing natural language processing (NLP) capabilities for Indic languages. Our mission is to bridge the gap in NLP research for low-resource Indic languages by providing high-quality datasets, pre-trained models, and tools tailored for diverse linguistic needs.

🚀 What We Do

• Datasets: Creation and publication of datasets for various NLP tasks, including translation, classification, and generation, with a focus on Indic languages.
• Models: Development of state-of-the-art NLP models fine-tuned for Indic languages, leveraging techniques like PEFT and LoRA.
• Research: Conducting and sharing research to solve key challenges in Indic NLP, including transliteration, low-resource learning, and domain-specific applications.

📚 Featured Projects

• Hellaswag-Telugu: A Telugu version of the Hellaswag dataset for advanced evaluation.
• Indic Language Translation and Transliteration: Custom tools and APIs for translation and mixed transliteration (Telugu-English).

🛠️ How to Contribute

We welcome contributions! Whether you’re interested in annotating data, building models, or sharing insights, feel free to get in touch.

🌐 Links

• Hugging Face Hub

📜 Citation

If you use our datasets or models in your research, please cite us as follows:

@misc{IndicVerse2024,  
  author = {Nikhil Chowdary Paleti and Divi Eswar Chowdary},  
  title = {Indic Verse: Datasets and Models for Advancing Indic Languages in NLP},  
  year = {2024},  
  publisher = {Hugging Face},  
  url = {https://huggingface.co./IndicVerse}  
}